190 



responding to 2 mm, before its effect is measured, and con- 

 sequently the effect of very soft y radiation may have been 

 missed. 



The lack of symmetry in quality of the emergence and 

 incidence radiation from some materials may be explained 

 if we suppose the distribution of the scattered radiation which 

 is produced from the hard y rays is not the same as for that 

 produced from the softer rays. The secondary radiation pro- 

 duced from the hard y rays appears to move on more in the 

 direction of the original stream than do the secondaries pro- 

 duced from the softer rays. 



The distribution of the secondaries may also to some ex- 

 tent depend upon the nature of the medium in which scat- 

 tering has taken place, and it does not seem at all unlikely 

 that the atomic structure of the radiator should determine to 

 some extent the nature and result of a collision between the 

 y particle and the constituent part of the atom. 



It has already been stated that these experiments give 

 little or no support to the theory of selective absorption ad- 

 vanced by Kleeman ; the effects seem much simpler than we 

 might be led to suppose from that theory. It seems possible, 

 also, to suggest probable causes for the effects which Klee- 

 man has observed. In some cases the absorbing screens used 

 by Kleeman seem not to be of sufficient thickness to pre- 

 clude the possibility of a certain amount of ^ rays, given off 

 by the radiator, penetrating the screen. Secondly, it has 

 been assumed that the incidence y radiation from all the sub- 

 stances experimented with is homogeneous. If, for example, 

 the incidence radiation from Pb contained a very soft bundle 

 in addition to the hard, many of the results obtained by 

 Kleeman could be immediately explained. With the present 

 form of apparatus I have not of course been able to test this 

 point ; it does not, however, seem at all improbable that Pb 

 should give out an exceedingly soft bundle of y rays, the 

 product of the soft y rays in the original radiation ; indeed, 

 if it is safe to make comparisons from the behaviour of such 

 substances as Zn, it would be surprising if such a soft bundle 

 did not exist, although of course its effect may be small. 

 Thirdly, in dealing with the very soft y rays, which consti- 

 tute the secondary incidence radiation, it seems necessary to 

 take into account the effect of tertiary radiation. 



In conclusion it may not be out of place to discuss one 

 or two points of some interest which have arisen during the 

 progress of these experiments. 



In fig. viii. it will be observed that for a mass of ab- 

 sorbing screen 70 the result of cutting out a considerable 

 amount of the secondary radiation has been to change the 



